Neutron spectrum unfolding using two architectures of convolutional neural networks - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Article Dans Une Revue Nuclear Engineering and Technology Année : 2023

Neutron spectrum unfolding using two architectures of convolutional neural networks

Résumé

We deploy artificial neural networks to unfold neutron spectra from measured energy-integrated quantities. These neutron spectra represent an important parameter allowing to compute the absorbed dose and the kerma to serve radiation protection in addition to nuclear safety. The built architectures are inspired from convolutional neural networks. The first architecture is made up of residual transposed convolution's blocks while the second is a modified version of the U-net architecture. A large and balanced dataset is simulated following "realistic" physical constraints to train the architectures in an efficient way. Results show a high accuracy prediction of neutron spectra ranging from thermal up to fast spectrum. The dataset processing, the attention paid to performances' metrics and the hyperoptimization are behind the architectures' robustness.
Fichier principal
Vignette du fichier
1-s2.0-S173857332300133X-main.pdf (1.36 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Licence : CC BY NC ND - Paternité - Pas d'utilisation commerciale - Pas de modification

Dates et versions

hal-04104489 , version 1 (24-05-2023)

Licence

Paternité - Pas d'utilisation commerciale - Pas de modification

Identifiants

Citer

Maha Bouhadida, Asmae Mazzi, Mariya Brovchenko, Thibaut Vinchon, Mokhtar Z Alaya, et al.. Neutron spectrum unfolding using two architectures of convolutional neural networks. Nuclear Engineering and Technology, 2023, 55 (6), pp.2276 - 2282. ⟨10.1016/j.net.2023.03.025⟩. ⟨hal-04104489⟩
78 Consultations
79 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More